A recently developed animal model for evaluation of anxiolytic and anxiogenic compounds involves two test batteries that elicit and measure defensive reactions to the presence of threat stimuli (Fear/Defense Test Battery), or to potential threat (Anxiety/Defense Test Battery). Present threat stimuli elicit flight, freezing, defensive threat and defensive biting, while freezing, risk assessment and the inhibition of nondefensive behavior are seen to partial predatory stimuli or places where predators have been seen. Because these tests do not use painful stimuli or involve learning of reactions to stimuli associated with pain, they are analytically less complex than traditional anxiety models, but provide a full range of defense behaviors for analysis of bidirectional change associated with administration of pharmacological compounds. Work completed with benzodiazepines, 5-HT ligands and imipramine suggests that drugs with clinical efficacy against anxiety produce specific patterns of change in risk assessment and inhibition of nondefensive behavior, found also with the NMDA antagonist MK-801, but not with morphine, scopolamine, or alprazolam. In contrast, the panickogenic agent yohimbine reduced freezing and enhanced flight and avoidance, suggesting a relationship between these behaviors and the target symptoms of panic attack. An additional feature emerging from this work is the consistent finding of gender differences in defensive behaviors and in sensitivity to anxiolytic drugs, in these rodent models. The proposed program will use these test batteries, plus tests of antipredator vocalization and analysis which differentiates anxiolytic and sedative profiles, to investigate patterns of effect for benzodiazepine receptor partial agonists, additional 5-HT ligands, and 3 classes of compounds with novel actions (CCK, CRF and NMDA ligands), including additional panickogenic agents. This program will provide a much finer analysis of the pharmacology of the neurobehavioral defense systems that has previously been available, and will additionally permit an analysis of gender effects in defensive behaviors linked to anxiety, and in reactions to different classes of anxiolytic drugs.